The apparent optical properties (AOPs) of oceanic case 1 waters were previously analyzed [Morel, 1988] and statistically related to the chlorophyll concentration ([Chl]) used as a global index describing the trophic conditions of water bodies. From these empirical relationships a bio-optical model of the upper layer was developed. With objectives and structure similar to those of the previous study the present reappraisal utilizes AOPs determined during recent Joint Global Ocean Flux Study cruises, namely, spectral attenuation for downward irradiance K-d(lambda) and irradiance reflectance R(lambda). This revision also benefits from improved knowledge of inherent optical properties (IOPs), namely, pure water absorption coefficients and particle scattering and absorption coefficients, and from better pigment quantification (via a systematic use of highperformance liquid chromatography). Nonlinear trends, already observed between optical properties and algal biomass, are fully confirmed, yet with numerical differences, The previous K-d(lambda) model, and subsequently the R(lambda) model, is modified to account for these new relationships. The R(lambda) values predicted as a function of [Chl] and the predicted ratios of reflectances at two wavelengths, which are commonly used in ocean color algorithms, compare well with field values (not used when developing the reflectance model). This good agreement means that semianalytical ocean color algorithms can be successfully applied to satellite data. Going further into purely analytical approaches, ideally based on radiative transfer computations combined with a suite of relationships between the IOPs and [Chl], remains presently problematic, especially because of the insufficient knowledge of the phase function and backscattering efficiency of oceanic particles.